Module 5 of 10

Advanced Prompting Techniques

System, role, contextual, and step-back prompting

🔧 System Prompting

System prompts set the overall behavior and context for the entire conversation. They act as global instructions that apply to all subsequent prompts.

Example Structure:

SYSTEM PROMPT:

"You are an expert financial advisor for Indian markets. Always provide advice in Indian Rupees, mention SEBI regulations when relevant, and consider tax implications for Indian residents."

USER PROMPT:

"Should I invest in mutual funds or stocks?"

The AI will now consider Indian context in all responses

✅ Best Use Cases:

Setting tone and personality (formal, casual, educational)
Defining domain expertise
Establishing output format preferences
Setting safety and ethical boundaries

👤 Role Prompting

Ask the AI to act as a specific persona or expert. This shapes the response style, vocabulary, and perspective.

💼 Professional Roles

"Act as a senior software architect..."

"Act as a legal advisor specializing in..."

"Act as a marketing strategist..."

🎓 Educational Roles

"Act as a tutor for a 10-year-old..."

"Act as a university professor..."

"Act as a career counselor..."

Real Example:

"Act as a travel guide for Indian tourists. Explain the visa process for visiting Europe in simple Hindi-English mix."

→ Response will use Hinglish, be informal, and include India-specific tips

📋 Contextual Prompting

Provide relevant background information to help the AI understand the specific situation and generate more accurate responses.

❌ Without Context:

"Write a job description for a software engineer."

Too generic - could be any company, any level

✅ With Context:

"Write a job description for a senior software engineer at a 50-person fintech startup in Bangalore. The role focuses on React and Node.js. We offer competitive salaries (15-25 LPA), ESOP, and remote work. Our culture is fast-paced and we value learning. Target candidates with 5+ years experience."

Specific, relevant, and actionable!

📌 Context Types to Include:

Target audience

Industry/domain

Constraints (budget, time)

Geographic location

Technical specifications

Company size/culture

🔄 Step-Back Prompting

Instead of diving into specifics, first ask the AI to reason about the broader principles, then apply them to your specific case.

Example Workflow:

Step Back Question:

"What are the general principles of good database design?"

Specific Question:

"Given these principles, design a database schema for an e-commerce platform."

✨ Why This Works:

Forces the AI to consider fundamental principles first
Reduces hallucinations and surface-level answers
Leads to more thoughtful, well-reasoned responses
Especially useful for complex technical or strategic questions

🎯 Combining Techniques (Pro Move!)

The real power comes from combining multiple techniques in one prompt.

Master Example (Uses All 4!):

SYSTEM

"You are an expert in Indian startup law."

ROLE

"Act as a legal advisor for a first-time entrepreneur."

CONTEXT

"My SaaS startup has 3 co-founders (40-30-30 split), raised ₹50L seed funding, based in Bangalore."

STEP-BACK

"First explain general principles of founder agreements, then create a checklist for our specific situation."

🎓 Key Takeaways

System prompts set global context for the entire conversation

Role prompting shapes personality, expertise, and communication style

Contextual prompting provides specific background for better accuracy

Step-back prompting ensures reasoning from first principles

Combining all four techniques creates professional-grade prompts

More context = more relevant and useful responses

Module 5 of 10 Complete